The Journal of Arthroplasty 30 (2015) 456–460

Contents lists available at ScienceDirect

The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org

Obesity, Morbid Obesity and their Related Medical Comorbidities are Associated with Increased Complications and Revision Rates after Unicompartmental Knee Arthroplasty Abdurrahman Kandil, MD ⁎, Brian C. Werner, MD, Winston F. Gwathmey, MD, James A. Browne, MD Department of Orthopaedic Surgery, University of Virginia Health System, Charlottesville, Virginia

a r t i c l e

i n f o

Article history: Received 29 July 2014 Accepted 11 October 2014 Keywords: obesity morbid obesity unicompartmental knee arthroplasty revision knee arthroplasty complications

a b s t r a c t Recent studies have demonstrated clinical success in expanding the indications for unicompartmental knee arthroplasty (UKA) to patients with increased body mass index (BMI). This study utilized national databases to identify 15,770 unique patients who underwent UKA between 2005 and 2011. 18.7% of patients undergoing a UKA were obese or morbidly obese. Univariate analysis demonstrated that obesity and morbid obesity were associated with significantly higher complication rates within 90 days postoperatively compared to non-obese patients. The overall short-term revision rate in obese and morbidly obese patients undergoing UKA was almost twice as high as the revision rate in non-obese patients. Obese and morbidly obese patients being considered for UKA should be counseled preoperatively regarding their increased risk of postoperative complications and revision surgery. © 2014 Elsevier Inc. All rights reserved.

Unicompartmental knee arthroplasty (UKA) is a valuable surgical option in patients with arthritis limited to a single compartment of the knee. UKA has several proposed advantages over total knee arthroplasty (TKA), including preservation of bone stock, shorter recovery, lower perioperative complications, improved function, and lower cost [1,2]. While knee arthroplasty utilization has increased over the past couple of decades, so too has obesity [3]. Not surprisingly, many patients undergoing total joint arthroplasty are obese and the effect of obesity on surgical outcomes is of interest to providers and patients. Body weight is a potentially modifiable risk factor for knee osteoarthritis (OA) and obesity increases the risk for the initiation and progression of knee OA [4–6]. In the studies of patients who have undergone total knee and hip arthroplasty, adverse events have been demonstrated to occur more commonly in obese patients, including dislocation, aseptic loosening, superficial infection, and revision surgery [7–10]. Less is known about the influence of obesity on UKA outcomes. Historically, the indications for UKA were older (N 60 years), lowerweight (less than 82 kg), lower-demand patients [11]. Multiple studies have corroborated these strict indications and have shown that increased weight leads to worse functional outcome scores, implant failure, and higher revision rates in patients undergoing UKA [12–14]. However, the indications for UKA are changing and many historical contraindications, such as weight less than 82 kg, are being challenged. Recent studies have demonstrated success in expanding the use of UKA to The Conflict of Interest statement associated with this article can be found at http://dx. doi.org/10.1016/j.arth.2014.10.016. Reprint requests: Abdurrahman Kandil, MD, Department of Orthopaedic Surgery, University of Virginia Health System, Ray C. Hunt Drive, Charlottesville, VA 22908. http://dx.doi.org/10.1016/j.arth.2014.10.016 0883-5403/© 2014 Elsevier Inc. All rights reserved.

heavier patients [15–17]. A recent study retrospectively reviewed 1000 mobile bearing UKAs and found no significant outcome difference, failure rate or survival between UKAs performed in ideal weight patients compared to non-ideal [16]. Another retrospective series of nearly 500 patients found that BMI greater than 30 did not affect clinical outcome or implant survivorship [18]. A retrospective review of 212 mobile bearing UKAs found that 10-year survival rates were similar between subgroups, whether stratified by BMI or weight [17]. While informative, all of these studies have been small and retrospective in nature and results have often been contradictory. To our knowledge, there is no large national database study looking at the effect of obesity or morbid obesity on postoperative complications in patients undergoing UKA. The goal of this study is to assess the association of obesity, morbid obesity and their related medical comorbidities with the risk of postoperative complications and need for revision surgery in patients undergoing UKA. Materials and Methods Data in this study were collected from the PearlDiver Patient Record Database (PearlDiver Inc, Fort Wayne, IN). The PearlDiver database is a publicly available, Health Insurance Portability and Accountability Act (HIPAA)-compliant national database. The database specifically has more than 2 billion individual patient records and contains Current Procedural Terminology (CPT) and International Classification of Diseases, 9th Revision (ICD-9) codes related to orthopedic procedures. There are two sub-databases within PearlDiver: a private-payer database and a Medicare-based database. The private-payer database has over 30 million individual patient records from 2007 to 2011. The

A. Kandil et al. / The Journal of Arthroplasty 30 (2015) 456–460

Medicare database has over 100 million individual patient records from 2005 to 2011. To prevent overlap of patients between the two databases or counting patients who used both private and Medicare insurances for a single procedure, all patients from the private-payer database were restricted to age less than 65 years and all patients from the Medicare database were restricted to age greater than 65. The database was queried for UKA using CPT code 27446 (arthroplasty, knee, condyle and plateau; medial OR lateral compartment). The dataset included both private payer (age b 65) and Medicare (age N 65) data. This search yielded procedural volumes, gender distribution, and age distribution for the CPT code. UKA patients were then divided into morbidly obese, obese and non-obese cohorts using ICD-9 codes for morbid obesity (BMI 40 and over): 278.01, V85.4, V85.41–V85.45, and obesity (BMI between 30 and 39): 278.00, V85.3, V85.30–V85.39. The morbidly obese, obese, and non-obese cohorts were queried for basic demographics including gender, age (b 65, 65–80, N80) and smoking status. Comorbidities for each cohort were assessed, including diabetes mellitus (DM), obstructive sleep apnea (OSA), hyperlipidemia (HLD), hypertension (HTN), peripheral vascular disease (PVD), congestive heart failure (CHF), coronary artery disease (CAD), chronic kidney disease (CKD), chronic lung disease, and chronic liver disease (CLD) using ICD-9 codes for each disease. Each cohort was then assessed for major complications within 90 days postoperatively, including pulmonary embolism (PE), deep vein thrombosis (DVT), postoperative infection, postoperative irrigation and debridement (I&D), myocardial infarction (MI), and respiratory failure. Each cohort was also assessed for minor complications, including postoperative stiffness, manipulation under anesthesia, (MUA), urinary tract infection (UTI), pneumonia, acute renal failure (ARF), cerebrovascular accident (CVA), acute cholecystitis, and need for blood transfusion. In addition, each cohort was assessed for local complications (postoperative infection, postoperative I&D, stiffness, MUA), medical complications (MI, respiratory failure, CVA, UTI, pneumonia, ARF), and separately for VTE (PE and/or DVT), infection (diagnosis and/or I&D), and stiffness (diagnosis and/or MUA). Due to potential HIPAA concerns, the PearlDiver database will not output patient numbers less than 11. Complications were grouped into “major”, “minor”, “local”, and “medical” categories to assure sufficient numbers were achieved to obtain output from the database. Additionally, overall revision rates, as defined by CPT codes, were compared between the morbidly obese, obese, and non-obese cohorts. Please refer to Table 1 for a list of all CPT and ICD-9 codes used in the study. Statistical comparisons were made with chi-square linear-by-linear association analysis with P b 0.05 considered significant. Chi-square tests, 95% confidence intervals, and odds ratios were calculated using SPSS, version 21. Results 15,770 unique patients underwent a UKA in the PearlDiver database over the 7-year period between 2005 and 2011. 2942 of these patients (18.7%) were coded as obese or morbidly obese. Obese and morbidly obese patients were more likely to be female and smokers (P b 0.0001). Obese and morbidly obese patients had higher rates of all assessed medical comorbidities, including DM, OSA, HLD, HTN, PVD, CHF, CAD, CKD, COPD, and CLD compared to non-obese patients (P b 0.0001). The highest associations were found between patients with or without obesity for the following comorbidities: DM [42.6 vs. 21.9%, respectively (P b 0.0001)] and OSA [20.5 vs. 4.5%, respectively (P b 0.0001)]. When compared to obese patients, morbidly obese patients had a higher rate of OSA (P b 0.0001). Interestingly, morbidly obese patients had a lower rate of HLD than obese patients (P b 0.0001). Aside from OSA and HLD, there are no significant differences in the remaining comorbidities between obese and morbidly obese. Table 2 provides a comparison of patient characteristics between the morbidly obese, obese, and non-obese cohorts.

457

Table 1 CPT and ICD-9 Codes Used in Study. Description Obesity and morbid obesity Morbid obesity (BMI 40 and over) Obesity (BMI between 30 and 39) Postoperative complications Pulmonary embolism (PE) Deep vein thrombosis (DVT) Diagnosis of postoperative infection Postoperative irrigation and debridement Acute myocardial infarction

Respiratory failure Stiffness requiring manipulation under anesthesia Urinary tract infection Pneumonia Acute renal failure Cerebrovascular accident

Acute cholecystitis Postoperative blood transfusion Revision knee surgery Revision of total knee arthroplasty, with or without allograft; femoral and entire tibial component Revision of total knee arthroplasty, with or without allograft; 1 component

CPT/ICD-9 Code ICD-9 278.01, V85.4, V85.41–V85.45 ICD-9 278.00, V85.3, V85.30–V85.39 ICD-9 415.1, 415.11, 415.19 ICD-9 453.4, 453.40, 453.41, 453.42 ICD-9 998.5, 998.51, 998.59, 711.9, 996.66, 996.69 CPT 23030, 23031, 23040, 29820, 29821, 29822, 29823 ICD-9 410.00-410.02, 410.10-410.12, 410.20-410.22, 410.30-410.32, 410.40-410.42, 410.50-410.52, 410.60- 410.62, 410.70-410.72, 410.80-410.82, 410.90-410.92 ICD-9 518.0, 518.51, 518.52, 518.81, 518.82 CPT 27570 ICD-9 098.0, 098.1, 098.10, 098.19, 599.0 ICD-9 480.0-480.9, 481, 482.0-482.9 ICD-9 584.5-584.9, 580.0-580.9, 586 ICD-9 430, 431, 432.0, 432.1, 432.9, 433, 433.0, 433.00, 433.01, 433.1, 433.10, 433.11, 433.2, 433.20, 433.21, 433.3, 433.30, 433.31, 433.8, 433.80, 433.81, 433.9, 433.90, 433.91, 434, 434.0, 434.00, 434.01, 434.1, 434.10, 434.11, 434.9, 434.90, 434.91 ICD-9 574.0, 574.3, 574.1, 575.0, 575.12, 574.4, 574.6, 574.8, 575.10 CPT 36430, ICD-9 V58.2, 990.0, 990.2, 990.3, 990.4 27487

27486

Obese and morbidly obese patients undergoing UKA had significantly higher major complication rates within 90 days postoperatively compared to non-obese patients (P b 0.0001). Obesity was associated with a two-fold greater risk of major complications (5.3 vs. 2.3%, respectively) (OR 2.3, 95% CI [1.9–3.0], P b 0.0001). Morbid obesity was associated with a three-fold greater risk of major complications (7.2 vs. 2.3%, respectively) (OR 3.2, 95% CI [2.5–4.2], P b 0.0001). Morbidly obese patients had a higher major complication rate than obese patients but this difference approached statistical significance (OR 1.4, 95% CI [1.0–1.9], P = 0.057) [Tables 3 and 4]. Minor complication rates within 90 days postoperatively were significantly higher in obese and morbidly obese patients undergoing UKA compared to non-obese patients (P b 0.0001). Obesity was associated with more than a two-fold greater risk of minor complications (9.8 vs. 4.1%, respectively) (OR 2.5, 95% CI [2.1–3.0], P b 0.0001). Morbid obesity was associated with more than a three-fold greater risk of minor complications (13.0 vs. 4.1%, respectively (OR 3.5, 95% CI [2.8–4.2], P b 0.0001). The difference in minor complication rates between morbidly obese and obese patients was statistically significant (OR 1.4, 95% CI [1.1–1.7], P = 0.012) [Tables 3 and 4]. While there was no difference in local complication rates between obese and non-obese patients, morbidly obese patients had significantly higher rates of local complications than obese or non-obese patients (6.7% vs. 3.7% or 3.4%, respectively) (P b 0.0001). Medical complication rates were higher in obese and morbidly obese patients compared to non-obese patients (10.4% and 7.8% vs. 2.0%, respectively) (P b 0.0001). In addition, morbidly obese patients had a significantly

458

A. Kandil et al. / The Journal of Arthroplasty 30 (2015) 456–460

Table 2 UKA: Patient Characteristics.

Total Number

Non-Obese

Obese

Morbidly Obese

12,928

1823

1019

Demographics Female Male Age b65 Age 65-80 Age N80 Smoker Comorbidities DM OSA HLD HTN PVD CHF CAD CKD COPD CLD

P Non-Obese vs. Obese

Non-Obese vs. Morbidly Obese

Morbidly Obese vs. Obese

6768 6160 6532 5162 1234 964

(52.4%) (47.6%) (50.5%) (39.9%) (9.5%) (7.5%)

1067 756 628 1049 146 186

(58.5%) (41.5%) (34.4%) (57.5%) (8.0%) (10.2%)

655 364 582 410 27 103

(64.3%) (35.7%) (57.1%) (40.2%) (2.6%) (10.1%)

b0.0001

b0.0001

0.003

b0.0001 b0.0001 0.039 b0.0001

b0.0001 b0.0001 b0.0001 0.003

b0.0001 b0.0001 b0.0001 0.988

2847 590 7357 8761 870 991 2657 838 1522 308

(22.0%) (4.6%) (56.9%) (67.8%) (6.7%) (7.7%) (20.6%) (6.5%) (11.8%) (2.4%)

778 328 1343 1514 209 314 632 243 418 112

(42.7%) (18.0%) (73.7%) (83.0%) (11.5%) (17.2%) (34.7%) (13.3%) (22.9%) (6.1%)

442 267 656 844 122 191 332 127 224 82

(43.4%) (26.2%) (64.4%) (82.8%) (12.0%) (18.7%) (32.6%) (12.5%) (22.0%) (8.0%)

b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 b0.0001

b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 b0.0001

0.748 b0.0001 b0.0001 0.920 0.731 0.334 0.278 0.548 0.595 0.064

DM: diabetes mellitus, OSA: obstructive sleep apnea, HLD: hyperlipidemia, HTN: hypertension, PVD: peripheral vascular disease, CHF: congestive heart failure, CAD: coronary heart disease, CKD: chronic kidney disease, COPD: includes chronic obstructive pulmonary disease and emphysema, CLD: chronic liver disease.

higher rate of medical complications as compared to obese patients (P b 0.0001) [Tables 3 and 4]. With regards to specific complications, obese and morbidly obese patients had higher rates of VTE than non-obese patients (2.2% and 2.1% vs. 0.8%, respectively) (P b 0.0001). While there was no difference in infection rates between obese and non-obese patients, morbidly obese patients had significantly higher rates of infection than obese and non-obese patients (3.8% vs. 1.6% and 1.2%, respectively) (P b 0.0001). There were no significant differences in stiffness between all three cohorts (P b 0.0001) [Tables 3 and 4]. The up-to 7-year revision rate in morbidly obese, obese, and nonobese patients undergoing UKA were 5.6, 4.6, and 2.7% (P b 0.0001). Morbidly obese patients had a 1.8-fold increased rate of revision than non-obese patients (P b 0.0001) whereas obese patients had a 2.2-fold increased rate of revision than non-obese patients (P b 0.0001). There was no difference in revision rates between morbidly obese and obese patients (P b 0.0001) [Tables 3 and 4].

Discussion Conflicting data exist regarding outcomes of UKA in obese and morbidly obese patients. Historically, weight greater than 82 kg was considered a contraindication to UKA due to concern for increased wear rates and an earlier need for a revision surgery [11]. In the present study, UKA patients who were obese and morbidly obese were more likely to have increased major, minor, local, and medical complication rates within 90 days postoperatively. In addition, obesity and morbid obesity were associated with significantly increased revision rates. It is interesting to note that aside from OSA and HLD, there are no significant differences in comorbidities between obese and morbidly obese, yet the morbidly obese have statistically higher minor, local, medical, and infection complication rates than obese patients. The prevalence of obesity has been increasing during the last decade of the 20th century3, and currently more than one-third of adults in the United States are obese [19]. Obesity has become a worldwide epidemic Table 4 90 Day Complication and Early Revision Rates.

Table 3 90 Day Complication and Early Revision Rates. Non-Obese Total number Major complicationsa Minor complicationsb Local complicationsc VTE (PE and/or DVT) Infection (diagnosis and/or I&D) Stiffness (diagnosis and/or MUA) Medicald Revision rate a b c d

12,928 303 2.3% 532 4.1% 439 3.4% 98 0.8% 160 1.2% 385 3.0% 256 2.0% 345 2.7%

Obese 1,823 97 5.3% 179 9.8% 68 3.7% 38 2.1% 29 1.6% 41 2.2% 142 7.8% 84 4.6%

Morbidly Obese 1,019 73 7.2% 132 13.0% 68 6.7% 22 2.2% 39 3.8% 31 3.0% 106 10.4% 57 5.6%

Major complications: PE, DVT, postop infection, postop I&D, MI, resp. failure. Minor complications: stiffness, MUA, UTI, PNA, ARF, CVA, acute chole, trans. Local complications: postop infection, postop I&D, stiffness, MUA. Medical complications: MI, respiratory failure, CVA, UTI, pneumonia, ARF.

Major complicationsa Minor complicationsb Local complicationsc VTE (PE and/or DVT) Infection (diagnosis and/or I&D) Stiffness (diagnosis and/or MUA) Medicala Revision rate a b c d

Morbidly Obese vs. Non-Obese

Obese vs. Non-Obese

Morbidly Obese vs Obese

OR [95% CI]

OR [95% CI]

OR [95% CI]

P

P

3.2 [2.5–4.2] P b 0.0001 3.5 [2.8–4.2] P b 0.0001 2.0 [1.6–2.6] P b 0.0001 2.9 [1.8–4.6] P b 0.0001 3.2 [2.2–4.5] P b 0.0001 1.0 [0.7–1.5] P = 0.984 5.7 [4.5–7.3] P b 0.0001 2.2 [1.6–2.9] P b 0.0001

2.3 [1.9–3.0] P b 0.0001 2.5 [2.1–3.0] P b 0.0001 1.1 [0.9–1.4] P = 0.506 2.8 [1.9–4.1] P b 0.0001 1.3 [0.9–1.9] P = 0.253 0.8 [0.5–1.0] P = 0.096 4.2 [3.4–5.2] P b 0.0001 1.8 [1.4–2.2] P b 0.0001

P 1.4 [1.0–1.9] P = 0.057 1.4 [1.1–1.7] P = 0.012 1.8 [1.3–2.6] P = 0.001 1.0 [0.6–1.8] P = 0.997 2.5 [1.5–4.0] P b 0.0001 1.4 [0.9–2.2] P = 0.244 1.4 [1.1–1.8] P = 0.022 1.2 [0.9–1.7] P = 0.284

Major complications: PE, DVT, postop infection, postop I&D, MI, resp. failure. Minor complications: stiffness, MUA, UTI, PNA, ARF, CVA, acute chole, trans. Local complications: postop infection, postop I&D, stiffness, MUA. Medical complications: MI, respiratory failure, CVA, UTI, pneumonia, ARF.

A. Kandil et al. / The Journal of Arthroplasty 30 (2015) 456–460

with wide medical, social, and financial implications. In our study, 18.7% of patients undergoing a UKA were coded as obese or morbidly obese. This is approximately half the rate of obesity nationwide. Patients under age 65 were more likely to be non-obese whereas those between the ages of 65 and 80 were more likely to be obese or morbidly obese. Obesity and morbid obesity are often associated with medical comorbidities that may independently affect outcomes. In our study, obese and morbidly obese patients were more likely to be smokers and have higher rates of multiple medical comorbidities than nonobese patients. This is in line with other studies, which have shown that obesity increases the risk of a number of health conditions including hypertension, hyperlipidemia, and type 2 diabetes mellitus [20–25]. Since our study is a univariate analysis, we have assessed the association of obesity, morbid obesity and their associated medical comorbidities with the risk of postoperative complications and need for eventual revision surgery and not the independent effect of obesity on these outcome measures. Body weight is a potentially modifiable risk factor for knee OA [4]. High BMI is related to structural damage of the articular cartilage and subchondral tibial bone thickening as there is a fourfold increase in joint forces for every kilogram of body weight gained [6]. Altered knee kinematics in obese patients may decrease the life of arthroplasty implants. It is thought that obesity leads to greater tibio-femoral loads and this excess weight translates to greater implant interface stress and increases the potential for early implant loosening [26]. This is supported by the findings in the present study that obese patients have nearly double the early revision rate as compared to non-obese patients. Multiple studies have demonstrated that increased weight leads to higher UKA revision rates. Bonutti et al compared 40 fixed-bearing UKA patients with a BMI less than 35 with 40 patients whose BMI was greater than 35 [12]. The authors observed an increased revision rate and poorer knee society scores in obese patients. Berend et al evaluated 79 UKAs with a mean follow-up of 40 months and observed that a BMI greater than 32 was an independent risk factor for implant failure and diminished survivorship [13]. Heck et al found that increased weight was associated with increased need for revision arthroplasty [14]. The present study specifically looked at within 7-year revision rates and did find a two-fold increased risk of revision for obese patients. The database has a 7-year dataset, thus revision data were only available for between 0 and 7 years postoperatively, depending on the date on which the procedure was performed. On the other hand, several recent studies have reported that increased weight may not be associated with higher revision rates. Pandit et al found no significant clinical or functional outcome difference, failure rate or survival between 551 UKAs performed in ideal weight patients (44–82 kg) compared to non-ideal (82–185 kg) [16]. Kuipers et al studied 437 Oxford 3 UKAs and found that BMI greater than 30 did not affect clinical outcome or implant survivorship [18]. Cavaignac et al conducted a retrospective review of 212 UKAs at mean 12-year follow-up and found that 10-year survival rates were similar between subgroups, whether stratified by BMI or weight. However, the authors noted that there was a non-statistically significant trend between weight and risk of revision [17]. Xing et al retrospectively reviewed a consecutive case series of UKA done by a single surgeon between 2004 and 2007 including 178 knees and found that the outcome of UKA was not influenced by the patient's BMI [27]. These findings are in contradiction to our study's findings. It should be noted that these studies had small numbers and may have been underpowered. A major strength of this study is the large sample size. To our knowledge, this is the largest nationwide database study to look at complications and revisions of obese and morbidly obese patients undergoing UKA. Our analysis includes Medicare and private payer data and captures the large number of younger (b65 years) patients undergoing UKA that has not been captured in some studies. Limitations of this study are due in large part to the nature of this database analysis. The PearlDiver database does not take into account

459

other preoperative factors or clinical concerns such as specific preoperative laboratory values, individualized data, and type of implant. Additionally, database studies rely on clinical coding and entering diagnoses, thus mis-coding or mis-diagnosing may alter results. In our study, revision rates are for revisions within the confines of the database, so the data are limited to early revisions only. In addition, this database does not specify laterality when determining revision. While unlikely, this limitation may over-estimate the revision rate in our study. Another limitation is our need for pooling complications to detect differences between groups as a result of low individual numbers and database limitations. This was done for major, minor, local, and medical complications. Finally, the PearlDiver database does not allow us to match obese, morbidly obese, and non-obese cohorts for medical comorbidities, age, and other potential confounding factors. The database only outputs groups of data and not individual data for each patient. We therefore are unable to perform a multi-variate regression analysis to determine if obesity or morbid obesity is independently associated with the complications listed. In conclusion, obesity, morbid obesity and their related medical comorbidities appear to be associated with higher rates of major, minor, local, and medical complications following UKA. In addition, early revision rates appear to be higher in obese and morbidly obese patients compared to non-obese patients. Obese and morbidly obese patients being considered for UKA should be counseled preoperatively regarding their increased risk of postoperative complications and revision surgery. References 1. Iorio R, Healy WL. Unicompartmental arthritis of the knee. J Bone Joint Surg Am 2003; 85-A:1351. 2. O'Rourke MR, Gardner JJ, Callaghan JJ, et al. The John Insall Award: unicompartmental knee replacement: a minimum twenty-one-year followup, end-result study. Clin Orthop Relat Res 2005;440:27. 3. Flegal KM, Carroll MD, Ogden CL, et al. Prevalence and trends in obesity among US adults, 1999–2000. J Am Med Assoc 2002;288(14):1723. 4. Gelber AC, Hochberg MC, Mead LA, et al. Body mass index in young men and the risk of subsequent knee and hip osteoarthritis. Am J Med 1999;107:542. 5. Niu J, Zhang YQ, Torner J, et al. Is obesity a risk factor for progressive radiographic knee osteoarthritis? Arthritis Rheum 2009;61:329. 6. Messier SP, Gutekunst DJ, Davis C, et al. Weight loss reduces knee-joint loads in overweight and obese older adults with knee osteoarthritis. Arthritis Rheum 2005; 52:2026. 7. Dowsey MM, Liew D, Stoney JD, et al. The impact of pre-operative obesity on weight change and outcome in total knee replacement: a prospective study of 529 consecutive patients. J Bone Joint Surg (Br) 2010;92(4):513. 8. Kim Y, Morshed S, Joseph T, et al. Clinical impact of obesity on stability following revision total hip arthroplasty. Clin Orthop Relat Res 2006;453:142. 9. Munger P, Roder C, Ackermann-Liebrich U, et al. Patient-related risk factors leading to aseptic stem loosening in total hip arthroplasty: a case–control study of 5,035 patients. Acta Orthop 2006;77:567. 10. Dewan A, Bertolusso R, Karastinos A, et al. Implant durability and knee function after total knee arthroplasty in the morbidly obese patient. J Arthroplasty 2009; 24(6 Suppl):89 [94.e1-3]. 11. Kozinn SC, Scott R. Unicondylar knee arthroplasty. J Bone Joint Surg Am 1989;71(1):145. 12. Bonutti PM, Goddard MS, Zywiel MG, et al. Outcomes of unicompartmental knee arthroplasty stratified by body mass index. J Arthroplasty 2011;26:1149. 13. Berend KR, Lombardi AV, Mallory TH, et al. Early failure of minimally invasive unicompartmental knee arthroplasty is associated with obesity. Clin Orthop Relat Res 2005;440:60. 14. Heck DA, Marmor L, Gibson A, et al. Unicompartmental knee arthroplasty. A multicenter investigation with long-term follow-up evaluation. Clin Orthop Relat Res 1993;286:154. 15. Borus T, Thornhill T. Unicompartmental knee arthroplasty. J Am Acad Orthop Surg 2008;16(1):9. 16. Pandit H, Jenkins C, Gill HS, et al. Unnecessary contraindications for mobile-bearing unicompartmental knee replacement. J Bone Joint Surg (Br) 2011;93(5):622. 17. Cavaignac E, Lafontan V, Reina N, et al. Obesity has no adverse effect on the outcome of unicompartmental knee replacement at a minimum follow up of seven years. Bone Joint J 2013;95-B(8). 18. Kuipers BM, Kollen BJ, Bots PC, et al. Factors associated with reduced early survival in the Oxford phase III medial unicompartment knee replacement. Knee 2010;17:48. 19. Hedley AA, Ogden CL, Johnson CL, et al. Prevalence of overweight and obesity among US children, adolescents, and adults, 1999–2002. J Am Med Assoc 2004;291:2847. 20. National Institutes of Health. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults—the evidence report. Obes Res 1998;6 (Suppl 2):51S. 21. Folsom AR, Kushi LH, Anderson KE, et al. Associations of general and abdominal obesity with multiple health outcomes in older women: the Iowa Women's Health Study. Arch Intern Med 2000;160:14.

460

A. Kandil et al. / The Journal of Arthroplasty 30 (2015) 456–460

22. Guh DP, Zhang W, Bansback N, et al. The incidence of co-morbidities related to obesity and overweight: a systematic review and meta-analysis. BMC Public Health 2009:9. 23. Hartemink N, Boshuizen HC, Nagelkerke NJ, et al. Combining risk estimates from observational studies with different exposure cutpoints: a meta-analysis on body mass index and diabetes type 2. Am J Epidemiol 2006;163:11. 24. Haslam DW, James WP. Obesity. Lancet 2005;366:9492.

25. Hu FB, Manson JE, Stampfer MJ, et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. N Engl J Med 2001;345:11. 26. Palumbo BT, Scott RD. Diagnosis and indications for treatment of unicompartmental arthritis. Clin Sports Med 2014;33(1):11. 27. Xing Z, Katz J, Jiranek W. Unicompartmental knee arthroplasty: factors influencing the outcome. J Knee Surg 2012;25(5):369.